Time-Dependent ROC Analysis

Time-Dependent ROC Analysis

Usage

timeroc(
  data,
  elapsedtime,
  outcome,
  outcomeLevel,
  marker,
  timepoints = "12, 36, 60",
  method = "incident",
  bootstrapCI = FALSE,
  nboot = 100,
  plotROC = TRUE,
  plotAUC = TRUE,
  timetypeoutput = "months",
  showOptimalCutoff = TRUE,
  showMarkerStats = TRUE,
  compareBaseline = FALSE,
  smoothAUC = FALSE
)

Arguments

data

.

elapsedtime

.

outcome

.

outcomeLevel

.

marker

.

timepoints

.

method

.

bootstrapCI

.

nboot

.

plotROC

.

plotAUC

.

timetypeoutput

Time units for display in plots and results.

showOptimalCutoff

Calculate and display optimal cutoff points that maximize Youden index.

showMarkerStats

Display descriptive statistics for the marker variable.

compareBaseline

Compare marker performance to a baseline model (AUC = 0.5).

smoothAUC

Apply smoothing to the AUC over time plot for better visualization.

Value

A results object containing:

results$text					a html
results$aucTable					a table
results$rocPlot					an image
results$aucPlot					an image
results$markerStats					a table
results$cutoffTable					a table
results$modelComparison					a html
results$clinicalInterpretation					a html

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$aucTable$asDF

as.data.frame(results$aucTable)